Hi there!

I am having issues determining the optimal data-driven bandwidth in a fuzzy RD design.

I am using the rdrobust package developed by Calonico et al 2016 (https://sites.google.com/site/rdpackages/home or https://rdrr.io/cran/rdrobust/man/rdrobust-package.html)

My dataset consists of paneldata with 400k+ obs.

The code I run is:

Code:
rdbwselect y center, c(0) fuzzy(z) kernel(triangular) vce(cluster dpnr) all
Where y = outcome, center = running variable, z = treatment status, dpnr = clustered ID variable

When I run the command, I get an error saying:

Invertibility problem in the computation of preliminary bandwidth below the threshold
Invertibility problem in the computation of preliminary bandwidth above the threshold
Not enough variability to compute the loc. poly. bandwidth (h) below the threshold. Range defined by bandwidth = .
Not enough variability to compute the loc. poly. bandwidth (h) below the threshold. Range defined by bandwidth = .
Not enough variability to compute the loc. poly. bandwidth (h) below the threshold. Range defined by bandwidth = .
Invertibility problem in the computation of bias bandwidth (b) below the threshold
Invertibility problem in the computation of bias bandwidth (b) above the threshold
Invertibility problem in the computation of loc. poly. bandwidth (h) below the thresholdInvertibility problem in the computation of loc. poly. bandwidth (h) above the threshold
I am currently using Stata/IC 15.1


I hope someone can help me out!

Best regards

Esben